According to the World Health Organization, corneal blindness from disease and injury is one of the top causes of blindness. Diabetes increases the risk of improper wound repair after minor scratches, corneal transplantation, and removal of epithelium during vitrectomy or refractive surgery. Altered focal adhesion, hemidesmosome, and cell-cell junction dynamics are common hallmarks of poor wound healing. Our group has demonstrated increased expression of the ATP-binding purinoreceptor, P2X7 in human diabetic corneas and in a murine model of Type II diabetes (DIO) accompanied by impaired wound healing. Furthermore, changes in P2X7 alters focal adhesion turnover and actin bundling in vitro that may lead to impaired wound healing. We have preliminary data from both wounded corneal epithelial cultures and explants demonstrating that sustained calcium oscillations occur after stimulation and travel between cells for hours. We found that when calcium mobilizations were absent between cells at the leading edge, that lamellipodial protrusions were absent. It is not known which receptor is required for this activity; however, we predict that the oscillations are a result of a highly-orchestrated activation of purinoreceptors. Previously we demonstrated that knockdown of specific purinoreceptors altered downstream signaling pathways that involve cell motility and adhesion proteins. Our goal in this innovative proposal is to use image processing and machine learning to understand the events that occur in cell-cell communication that induce cell migration in corneal epithelia and to examine how these events are mediated in corneas from a diabetic induced obesity murine model. The proposal challenges the accepted paradigm that the cell-cell communication wave in the wounded cornea is a short-lived on-and-off signal, but that it is a prolonged oscillating signal needed for healthy epithelial cell migration to effectively heal the wounded cornea. As the calcium and motility assays yield large data sets and complex patterns, we are collaborating with a neuroscientist and a computer scientist to utilize machine learning to determine when cells will interact if specific receptors or complexes of receptors are activated. To address these questions, we will simultaneously perform live cell imaging of calcium mobilization, actin and ATP release in epithelium of wounded corneas under normal and diabetic conditions. The aims are: 1. To determine if there are specific response signatures elicited by the sustained calcium mobilizations and signaling generated with injury in control corneas and corneas from diet induced obesity mice; and 2. To determine if the intercellular calcium mobilizations required for the initiation of lamellipodial ruffling are necessary for proper migration?